International Journal of Infectious Diseases 51 (2016) 128132

Contents lists available at ScienceDirect

International Journal of Infectious Diseases

journal homepage: www.elsevier.com/locate/ijid

Higher incidence of Zika in adult women than adult men in Rio de

Janeiro suggests a signicant contribution of sexual transmission from
men to women
Flavio Codeco Coelho a,*, Betina Durovni b, Valeria Saraceni b, Cristina Lemos b,
Claudia Torres Codeco c, Sabrina Camargo a, Luiz Max de Carvalho c,d, Leonardo Bastos c,
Denise Arduini b, Daniel A.M. Villela c, Margaret Armstrong a,e
a

Fundacao Getulio Vargas, Praia de Botafogo, 190, Rio de Janeiro, Brazil

Secretaria Municipal de Saude, Rio de Janeiro, Brazil
c
Programa de Computacao Cientca, Fundacao Oswaldo Cruz, Rio de Janeiro, Brazil
d
Institute of Evolutionary Biology, University of Edinburgh, Edinburgh, UK
e
MINES Paristech, PSL Research University, CERNA Centre for Industrial Economy, i3, CNRS UMR 9217, Paris, France
b

A R T I C L E I N F O

Article history:
Received 24 June 2016
Received in revised form 16 August 2016
Accepted 27 August 2016
Corresponding Editor: Eskild Petersen,
Aarhus, Denmark
Keywords:
Zika
Sexual transmission
Dengue
Age-adjusted incidence

S U M M A R Y

Objectives: The recent emergence of Zika in Brazil and its association with an increased rate of congenital
malformations has raised concerns over its impact on the birth rate in the country. Using data on the
incidence of Zika in 20152016 and dengue in 2013 and 20152016 for the city of Rio de Janeiro
(population 6.4 million), a massive increase of Zika in women compared to men was documented.
Methods: The age-adjusted incidence was compared between men and women. A negative binomial
Poisson generalized linear model was tted to the Zika incidence data to determine the signicance of
sexual transmission statistically.
Results: Even after correcting for the bias due to the systematic testing of pregnant women for Zika, there
were found to be 90% more registered cases per 100 000 women than men in the sexually active age
group (1565 years); this was not the case for age groups <15 years and >65 years. Assuming that
infected men transmit the disease to women in their semen, but that the converse is not true, some extra
incidence in women is to be expected. An alternate hypothesis would be that women visit doctors more
often than men. To test this, the incidence of dengue fever was compared in men and women in 2015 and
in 2013 (before Zika reached Rio de Janeiro): in both years, women were 30% more likely to be reported
with dengue.
Conclusion: Women in the sexually active age group are far more likely to get Zika than men (+90%
increase); sexual transmission is the most probable cause. Women in the 1565 years age group are also
30% more likely to be reported with dengue than men, which is probably due to women being more
careful with their health.
2016 The Author(s). Published by Elsevier Ltd on behalf of International Society for Infectious Diseases.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/bync-nd/4.0/).

1. Introduction
Viral diseases transmitted by Aedes aegypti mosquitoes, such as
dengue, yellow fever, chikungunya, and Zika, have traditionally
been restricted to the tropical regions of the world, given the
intolerance of the vectors to colder climates.1 In these regions

* Corresponding author.
E-mail address: fccoelho@fgv.br (F.C. Coelho).

transmission tends to be modulated by temperature, slowing

down signicantly when temperatures drop below 20 8C. The
trend in global warming has long been argued to be a threat to
public health, as this will extend the reach of tropical diseases.13
Preparedness for these diseases is the order of the day for the
health agencies of countries with a temperate climate.
The emergence Zika as a global pandemic threat is changing the
traditional risk scenarios. The Zika virus has the ability to infect
other species of mosquitoes,4,5 thereby potentially extending its
reach. Aedes albopictus, for instance, is well established in

http://dx.doi.org/10.1016/j.ijid.2016.08.023
1201-9712/ 2016 The Author(s). Published by Elsevier Ltd on behalf of International Society for Infectious Diseases. This is an open access article under the CC BY-NC-ND
license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

F.C. Coelho et al. / International Journal of Infectious Diseases 51 (2016) 128132

temperate climates. Moreover, the Zika virus can also be

transmitted directly from human to human.6 The sexual route
seems to be the most common alternate form of transmission, but
the virus is also present in other bodily uids such as saliva and
urine.79
The recent arrival of Zika in Brazil in 2014, and the speed with
which it spread throughout the country and into neighboring
countries in just a few months, seems to indicate alternate forms of
transmission.
In this study, the age-adjusted incidence of Zika compared to
dengue in the city of Rio de Janeiro was analyzed and the relative
importance of the sexual route of transmission to the overall
incidence of Zika was estimated. By considering this additional and
asymmetric route of transmission, it was expected that a higher
incidence of Zika would be seen in women in the sexually active
age group.

2. Methods
The data used in this analysis were obtained from the Rio de
Janeiro health secretariat, and consist of every notied case of Zika
and dengue for the years 2013 (dengue only), 2015 (Zika and
dengue), and 2016 (Zika and dengue, up to April). Each record
includes the date of notication, International Classication of
Diseases tenth revision (ICD-10) code, age in years, sex, and
gestational status.
The ofcially estimated population of Rio de Janeiro for 2015,
based on the 2010 census, was also used (6.4 million people:
3 million men, 3.4 million women; Figure 1). All age-adjusted
incidences used in this study were calculated using this population
as standard.
The cases were aggregated to the same age classes, as shown in
the citys age pyramid (Figure 1). The classes are 5 years wide and
incidence values are presented as the number of cases per
100 000 inhabitants.

129

The incidence of Zika in women was calculated with and

without pregnant women, to avoid biases. The city health services
have systematically tested pregnant women displaying a skin rash,
due to the high risk of babies developing neurological complications caused by intrauterine Zika virus infection.
To check the statistical signicance of the increase in incidence
observed in women, a Poisson generalized linear model (GLM) was
rst tted to the number of cases of Zika and dengue (2013)
aggregated by age class. Since these initial models displayed
signicant over-dispersion, a negative binomial GLM was tted to
the same data. Let Y be the number of cases; the nal model is then:
Y i  NegBinpi ; r
pi r=r mi
logmi b0 b1 Sexi b2 Activei b3 Pregnanti

(1)

b4 Sexi Activei
The Active dummy variable takes the value 1 for age classes
above 15 years and less than 65 years of age, and 0 otherwise.
3. Results
During the recent Zika epidemic in Rio de Janeiro, which started
in late 2015, a total of 29 301 cases, 20 315 women and 8986 men,
were notied as suspected Zika cases based on clinical assessment.
During the same period, 102 754 total cases of dengue were
notied, with 46 305 being men and 56 449 being women. The
incidence by age group for Zika in the period January 2015 to April
2016 is shown in Figure 2. After removing the pregnant women
from the dataset, the incidences shown in Figure 3 were obtained.
The age distribution of pregnant women removed from the sample
is shown in the Supplementary Material (Figure S1).
For comparison, Figure 4 shows the age distributed incidence of
dengue in the 20152016 period. Note that the extra incidence in
women is far less pronounced. The dengue incidence in 2013 was

Figure 1. Age pyramid for the city of Rio de Janeiro. These numbers are ofcial projections based on the census of 2010. The male population is represented in blue, and the
female population in red.

130

F.C. Coelho et al. / International Journal of Infectious Diseases 51 (2016) 128132

Figure 2. Incidence of Zika in men and women by age group. The incidence is in units of cases per hundred thousand.

also investigated (Figure 5), to make sure that the pattern was not
specic to the 2015 dengue epidemic.
The combined incidence in the sexually active age group, from
15 years to 65 years of age, is shown in Table 1, along with the ratio
of the incidence in women to the incidence in men in the same age
group.
The regression results indicated a signicantly higher Zika
incidence for sexually active women (1.7767, 95% condence
interval (CI) 0.500 to 3.053, p = 0.006). Sex alone was not a
signicant predictor of Zika incidence (0.2120, 95% CI 1.207 to
0.783, p = 0.676). For dengue, being in the sexually active age group
and being a woman was not a signicant risk factor (0.7196, 95% CI
0.321 to 1.761, p = 0.138). Again, sex alone did not prove to be a
signicant risk factor for dengue.

4. Discussion
If Zika virus is being transmitted both through vectors and
sexually in Brazil and other American countries in this recent
epidemic, it is important to estimate the relative importance of
each route. According to the available evidence, the principal way
to transmit Zika virus sexually is through exposure to infected
seminal uid. Female to male sexual transmission of Zika virus has
been reported,10 but no information is available about how likely
such a transmission is. Assuming heterosexual intercourse to be far
more prevalent than homosexual sex between men, a surplus of
Zika cases in women due to sexual transmission can be expected.
Moreover active viruses have been isolated from semen more than
3 weeks after the onset of symptoms,9 which greatly increases the

Figure 3. Incidence of Zika in men and women by age group, excluding pregnant women. Pregnant women are excluded because extra efforts were made by the health
services to identify all possible Zika cases in this group due to their babies being at high risk of developing neurological complications.

F.C. Coelho et al. / International Journal of Infectious Diseases 51 (2016) 128132

131

Figure 4. Age-adjusted incidence of dengue in men and women for the 20152016 period.

transmission period for susceptible females. On the other hand,

without signicant sexual transmission and in the absence of an
observational bias favoring women, an equal incidence across age
groups would be expected.
What was observed, however, was a markedly higher incidence
in women, as shown in Figure 2. As mentioned in the Methods
section, there is at least one observational bias towards reporting
women cases, due to concerns about microcephaly and the risks to
babies. In order to minimize this, pregnant women were removed
from the sample (Figure 3). What is seen is that the extra incidence
remains, but is relatively smaller. However, this extra incidence is
more pronounced in the reproductive age group; these women are
more likely to visit a physician regularly and thus Zika will be
detected if they are infected.

To conrm the hypothesis that women are more likely to be

diagnosed with exanthematic fever syndromes during their
reproductive years, dengue incidence was assessed by age group
(Figures 4 and 5). During the 20152016 period, it is possible that
some Zika cases may have been misdiagnosed as dengue, thus
contaminating the dengue incidence with sexual transmission.
Thus the incidence of dengue in 2013 was also assessed, when Zika
was unlikely to be circulating in Rio de Janeiro. It can also not be
discounted that there is evidence that pregnant women are more
likely to develop severe dengue.11 Other factors that could explain
the higher incidence in women are (1) women are more likely to
stay at home and be more exposed to the vector, and (2) women of
a fertile age are more predisposed to see a doctor as soon as they
develop symptoms for fear of complications in an as yet

Figure 5. Age-adjusted incidence of dengue in men and women for the year 2013. Notice the pattern is similar to that observed in 20152016.

F.C. Coelho et al. / International Journal of Infectious Diseases 51 (2016) 128132

132

Table 1
Aggregated incidence in men and women in the sexually active age group of 1565
years. The last column shows the ratio of the incidence in women to the incidence in
men.
Disease

Incidence in
women

Incidence in
men

Ratio,
women/men

Zika
Dengue 201516
Dengue 2013

5382.58
16 628.04
32 201.37

2854.85
13 941.05
25 410.52

1.88
1.19
1.27

undetected pregnancy. The authors believe that the 2013 dengue

data serve as a good control for the differential exposure of women
to the vector. As for the behavioral changes associated with the fear
of having a baby with microcephaly, the 20152016 data on
dengue also represent a good control for this, due to the similarity
of the symptoms of dengue and Zika.
As can be seen in Table 1, the extra detection in women due to
the behavior of seeing a doctor regularly could account for at most
a 30% higher incidence than in men, if all the extra incidence of
dengue in women is attributed to this factor. However, for Zika,
even discounting pregnant women, an incidence that is almost 90%
higher than for men is observed. This extra incidence can be
attributed to the extra cases caused by sexual transmission.
Through sexual transmission, Zika is no longer constrained to
tropical and subtropical regions and will be able to reach northern
Europe, the northern USA and Canada, and northern Australia, as
well as Japan and Korea. Although it would be harder for the
disease to invade these higher latitudes, the incidences are likely to
be higher as men who catch Zika abroad can transmit locally for
weeks or even months.
The immediate consequence of this higher incidence of Zika in
women in the reproductive age group is a much higher expected
number of neurologically compromised babies than if the disease
was only transmitted through vectors. Going further, women living
in Zika-infested areas will think twice about falling pregnant, at
least those with access to birth control. This could well lead to a
drop in the birth rate, particularly in the middle classes.
As the Aedes mosquito is known to be present in the southern
states of the USA around the Gulf of Mexico,3 in southern Europe,2
and in northern Australia, an outbreak could start because of a
returning traveler, especially if sexual transmission propagates it
as well. Health authorities in developed countries are already
warning travelers visiting Zika-infected areas to consider delaying

pregnancies.12,13 What would happen to the birth rate in these

countries if Zika became endemic? A drop in births in Europe could
have serious economic consequences.14
Funding: None.
Conict of interest: None.

Appendix A. Supplementary data

Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.ijid.2016.08.023.
References
1. Brady OJ, Golding N, Pigott DM, Kraemer MU, Messina JP, Reiner Jr RC, et al.
Global temperature constraints on Aedes aegypti and Ae. albopictus persistence
and competence for dengue virus transmission. Parasit Vectors 2014;7:14.
2. Schaffner F, Mathis A. Dengue and dengue vectors in the WHO European region:
past, present, and scenarios for the future. Lancet Infect Dis 2014;14:127180.
3. Kraemer MU, Sinka ME, Duda KA, Mylne AQ, Shearer FM, Barker CM, et al. The
global distribution of the arbovirus vectors Aedes aegypti and Ae. albopictus. Elife
2015;4:e08347.
4. Ledermann JP, Guillaumot L, Yug L, Saweyog SC, Tided M, Machieng P, et al.
Aedes hensilli as a potential vector of chikungunya and Zika viruses. PLoS Negl
Trop Dis 2014;8:e3188.
5. Wong PS, Li MI, Chong CS, Ng LC, Tan CH. Aedes (Stegomyia) albopictus (Skuse): a
potential vector of Zika virus in Singapore. PLoS Negl Trop Dis 2013;7:e2348.
6. Mansuy JM, Dutertre M, Mengelle C, Fourcade C, Marchou B, Delobel P, et al.
Zika virus: high infectious viral load in semen, a new sexually transmitted
pathogen. Lancet Infect Dis 2016;16:1389.
7. Musso D, Roche C, Nhan TX, Robin E, Teissier A, Cao-Lormeau VM. Detection of
Zika virus in saliva. J Clin Virol 2015;68:535.
8. Gourinat AC, OConnor O, Calvez E, Goarant C, Dupont-Rouzeyrol M. Detection
of Zika virus in urine. Emerg Infect Dis 2015;21(1):846.
9. DOrtenzio E, Matheron S, de Lamballerie X, Hubert B, Piorkowski G, Maquart M,
et al. Evidence of sexual transmission of Zika virus. N Engl J Med
2016;374(22):21958.
10. Davidson A, Slavinski S, Komoto K, Rakeman J, Weiss D. Suspected female-tomale sexual transmission of Zika virusNew York City, 2016. MMWR Morb
Mortal Wkly Rep 2016;65:7167.
11. Machado CR, Machado ES, Rohloff RD, Azevedo M, Campos DP, de Oliveira RB,
et al. Is pregnancy associated with severe dengue?. A review of data from the
Rio de Janeiro surveillance information system.
PLoS Negl Trop Dis
2013;7:e2217.
12. Ahmad SS, Amin TN, Ustianowski A. Zika virus: management of infection and
risk. BMJ 2016;352:i1062.
13. Vouga M, Musso D, Van Mieghem T, Baud D. CDC guidelines for pregnant
women during the Zika virus outbreak. Lancet 2016;387:8434.
14. Kassam A, Scammell R, Connolly K, Orange R, Willsher K, Ratcliffe R. Europe
needs many more babies to avert a population disaster. UK: The Guardian;
2015 , Available at: http://www.theguardian.com/world/2015/aug/23/
baby-crisis-europe-brink-depopulation-disaster (accessed 2016-05-04)